Pegasystems Inc. (PEGA) Earnings Call Transcript & Summary

We have Pegasystems here at the Citizens JMP Technology Conference being held at the Ritz-Carlton Hotel in San Francisco. Ken and Alan, thank you guys so much for coming. The -- our road map for today is that we will talk, initially, sort of the standard house business. And then Pega has actually a super interesting new product called Blueprint, which we're going to talk about a little bit. And then we'll go into -- if we have any time left, we'll go into more standard things, but I definitely want to give a little bit of time to Blueprint.

Okay. So starting at the top, who wants to answer how's business?

I think, we closed the year very strong. We were really quite pleased. The teams came through broadly. Customers showed renewed buying interest, which is always awesome when that happens. You never take that for granted. And so I felt like we were coming into 2024 with some wind in our sails, both because of just the market not having collapsed, like some people were afraid it was going to, but also because of our product portfolio, which, in the first half of the year, is going to sort of culminate in releases as we enter PegaWorld in June.

Yes. All right. Well, let's touch on each of those. So let's touch on the market didn't collapse the way some people thought it was.

Well, look, the whole debate, I can remember a year ago, the debates on hard landing, soft landings, no landings. And I don't know how you want to categorize it, but buying behavior continues to exist. Customers are selective. It is definitely, definitely a constrained environment. Customers are very worried about what they're spending their money on, but they are willing to make investments, even if it's a little slower, even if it's a little more cautiously so...

Yes. How did that -- how was the arc of that sort of customer buying behavior last year, and I mean, Q1, Q2, Q3, Q4 ahead?

I think...

So we had a little bit of the middle of the year, I would say Q2 into Q3, one of the fears that, I think, a lot of companies were reacting to and quite frankly, the investment community was probably more reacting to this than our clients, but nonetheless, was gen AI and what's that going to do? And oh my gosh, companies are going to go out of business, and everybody is going to be -- when is your next renewal because you're not renewing your ERP system because gen AI is going to do all your accounting. Like everything was just kind of like this irrational -- and I do think that there were -- some of our clients where they're buying patterns were maybe distracted, at least, for us, a little bit into Q2 and the beginning of Q3. And then -- but it settled pretty quickly. And I think people started realizing like this gen AI stuff is real, but it also isn't like what everybody kind of played it out to be early on. Like it's forming into different use cases than what I think people originally thought. So I do think -- so our arc was a little bit of Q1. There's always a little bit of the kind of flow over from the Q4 previous. Q4s are always a big event with our clients so that gives us a selling opportunity. Q2 was, I would say, a little softer, and that probably played into the first part of Q3.

I've already forgotten about that whole...

March.

Everyone had to be the categorized as an AI winner or an AI loser.

It was March 21st when the first basket was produced. Thankfully, we weren't in the you're s****** basket, but there were some other companies that were.

Oh, my God. [indiscernible] had a really bad earnings call. That's right. Okay. Okay. So that's good. So that was the -- that was sort of the buying behavior side of it. The other part of it, how you're saying, that makes you feel good about this year is the product portfolio. So before you do Blueprint, you want to give us a big picture on the product portfolio?

Well, I think if you go talk to clients, you quickly hear that all they are hearing from vendors is all AI, all the time. And the level of -- it gets to the point where the SEC even talks about AI and whitewashing, right, or don't AI wash what you're saying. So you know there's a lot of hype in the market. But the thing that's very true here is there are extremely real and genuine drivers of value that come out of the proper application of some of this AI. So the first thing I'd say is there's a lot of confusion. We've been in the statistical AI world for well over a decade, and we still are. Statistical AI is where you do machine learning, figure out the next-best action for a customer, make them an offer, retain them, do those types of things. And in the bubble of AI, statistical AI and gen AI are getting confused even though they are fundamentally different, both in terms of technology and also candidly in terms of the use cases you would apply them on. But the statistical AI sets us up really well for being able to...

I had to go buy a book. Do you know this guy, Stephen Wolfram?

I do.

I figured you might.

Yes. And that's actually a good book. It's only massively oversimplified.

But that's what makes it so good for me. I'm on Page 2, and I've already learned a lot. So I didn't understand what temperature was.

Yes.

In the first 2 pages, he understands temperature, which is just how creative do you want your LLM to be, right? In the second...

How much hallucination...

Right. Yes. And anyways, so, no, I think your point is well taken.

It is a good book, and he actually has his own search engine, which if you haven't tried. You should just Google his name and the Wolfram search engine is actually very powerful and very good for certain types of problems, particularly ones that you think of as mathematically oriented.

Yes. I'm making a note of it. Okay. But you're right. I totally strayed. Your point was that statistical AI has been going for at least a decade, right?

Decade, yes.

And generating results, right? And then the gen AI stuff comes along and there's sort of a lot of -- and you can hear it in the conversations with the management teams, right? AI is used in a...

It's a [indiscernible]

Very broad. Yes, yes, yes.

And they were very -- also very different use cases between the use cases in which you would "train your own model," or the use cases in which you would use other techniques with gen AI to get output. And I'm happy to say, I think we understand this stuff extremely well. We invested the effort, had [indiscernible] people. I personally am very deeply involved in trying to understand and discuss with our thought leaders what we're doing. And I think our strategy here is terrific. So if I was going to summarize that for you, a lot of people have said, look, gen AI, which has the ability to summarize conversations, has the ability to generate a letter, has the ability to write better than 95% of the people in our companies, it's enormously powerful. And there's a set of features like summarization, generation, chatbot-type response, some of those features which are great features to use and to apply. And we've got about 22 of those that we've already added to our product to do important things that a customer would find to make themselves more feature rich as they work with their clients. But we actually came up with another way to apply gen AI that's profoundly different and that I don't see anybody else do it. And that other way is to say, all right, love the features, but what if we could turn the power of gen AI onto the very process fabric of a business itself into the very way that it thought of onboarding a customer, selling a product, doing a follow up. What if we could get the gen AI to be able to influence the workflows of the business? And of course, we're in the AI decisioning and workflow business. So we have this decisioning capability, largely statistical, and a workflow capability. And what we realized is that it would be possible to ask a prospect or a customer a certain number of structured questions, allow them to just go free text, pontificate, put in their business objectives and the other sorts of things they want to do in here as well. And then we could take that. We could go out to the Internet and find the best practices in the Internet and organize those for the particular process in mind. We could then take our best practices from the 4 decades that we've been building systems. And we could then also use gen AI, first gen AI to get the information on the Internet, but then use gen AI to extract value from our processes, and then use gen AI to marry those 2 together so that you get a synthesis of how you want to do something. What are the steps and stages that you want to go through if you're doing a KYC, a know-your-customer evaluation. And the quality of what comes out from this is absolutely mind blowing. It is -- well, I had a CEO of a U.K. bank come by a booth that we were in. We just rolled this out. And he said, putting in, how should I have my bank eliminate the account of a politically exposed person. Now the reason this was sensitive is Coutts Bank had gotten rid of a prominent liberal politician in the U.K. and had led to the forced resignation of the CEO because the process wasn't done exactly right. What this thing came up with was, he looked at us and says, that's brilliant. It had the right checks about compliance checks and consider the political ramifications and other types of things. So being able to take the Internet's input to process optimization, being able to take our stuff as input, and that's just the beginning because I'll tell you, the next 2 things we're doing between now and gen AI.

All right. We're talking about Blueprint now, right?

This is Blueprint.

All right. So just for everyone, now we're talking about Blueprint. And Blueprint is GA or no?

So Blueprint is GA with -- the first version of Blueprint is GA with our partners. So we decided to open it up to our partners like Accenture, Cognizant, whatever, so that they could get really comfortable with it this month. And at the end of this month, we're opening it up to, not just our customers, but to prospects as well.

Okay. So we're at the beginning of March, right, so Blueprint generally available to partners in March. And then in April, it is going to be generally, generally available, right?

Yes.

Okay. And then...

And I'll...

Don't take this the wrong way if people didn't understand it, okay. So I saw at, whatever we did, a 0.5 hour, 45-minute demo. When you see the demo, it really clicks. And I think it might be hard to make it click with the words. So let me just ask. Did it click? Do people get it? Or do we need an example? And I'm looking at you 3 because you can all answer. So awesome. You saw the demo? Oh, you didn't see the demo, right? Okay, did it click? It's okay to say no, it didn't click.

No. [indiscernible]

Okay. Yes. So...

So we can help you out. If you go to pega.com.

Let's go back to an example. I think the example is really powerful.

Okay. So if you think about, say, how you handle, you're a bank...

You're a bank.

And you handle a deceased customer. Sensitive situation, right? Actually, it turns out it's more complicated than people realize. You have to close accounts. You have to hold certain things in the [indiscernible]. You have to do distributions. You'd like to have a stage in there in which you retain the money, right? You don't want -- you don't want the family to pull the money out of the bank. Is there a chance that you can retain them? But boy, you have to handle that gingerly if you're going to do it. What Blueprint will do with [ face, ] with a description of just simple verbal description of -- I'm a retail bank or a high-net worth individual bank. How do I handle the process of managing and closing the accounts of a deceased customer. So how do you put it?

All right. So we had typed in. We are this is super regional bank. We have a private banking arm. How do we handle deceased customers in the banking arm? It's actually super relevant because of the whole transfer from, right? Big deal right now, yes.

And it will give you a little guidance as you're typing it. And if you don't like what you've typed, you can just retype it, because the thing about this is it's not driven by what the computer system needs. The power of gen AI means it's driven by how you choose to explain it. And by the way, if you wanted a version of this in Spanish, you just click a little button that says Spanish. I was in Germany last week, and we were popping these things out in German. And I know I do what they said, but the audience was really impressed. So you put that in and what comes out of it is the first thing that comes out and say, here are the types of things you need to do, right? You're going to have to go through an account closure process here as well. You're going to have to have a notification process. You're going to have a transfer of assets process. You may have a payment of debt process. It's going to tell you what are the workflows that you need. Then it will dig into each one of those workflows, and it will tell you this workflow consists of the following 5 steps: Go out and get the balances from all of the accounts; put a stop on all of the accounts; and more doesn't come out here. Sometimes you have to wait 3 days for transactions to clear, right? And then to all of those steps, including, is this an automated step? Is this a step that involves a person? Is this a step that involves a system? Then the next thing it does is it says, here's the data that's going to be needed for each of these stages of these steps. And this is where in our next cut, we're going to make it so you can load in your data sources. So we actually tell you using your own data how you want this process to run. The next thing it shows you with pretty little pictures is here are all the different personas. Here's all the different people involved. There might be a compliance officer involved, right? There's the account representatives. So it will actually tell you everybody that's there. And then it will create a document for you, a PDF that you can give to your compliance team, and say, this is how we plan to run the deceased account closure process. Do you like it? If they don't like it, it's easily changeable. So take something to be blunt, it used to take months and you actually get a challenger approach to design thinking in a couple of minutes. So -- and it's beautiful. Now the next bonus is there are 2 buttons at the bottom, only one of which works today. One is generate PDF. And the second is feed into your Pegasystems. And if you're a customer, if this is a customer who has a Pegasystems...

And you guys can't generate the PDF yet.

No PDF.

No, we get it.

PDF is [indiscernible] Actually, as of this week, we're getting the Pega generating...

So that's going to start generating -- but it won't be your final version of your product, obviously. But this will start -- well, I don't know, maybe this is the final version of your product.

Well, it's a pretty -- so there are 2 things that are going on. One, our tagline as a company is built for change. So the idea that there's a final version mean things aren't changing...

You're right. Okay. It's adaptable.

This stuff changes, life changes this stuff all the time. But we are going to continue to build out all of the other elements that will make this a more and more fulsome product. And the thing I told my customers -- we have PegaWorld in June. And all of you should come to PegaWorld in June, June 9 to 11 in Las Vegas this year. They have an investor session -- yes, Ken -- as well.

Monday, the 10th?

Is that I expected that a year from now, so this coming June, based on the work we were doing with AI, that we would be able to double the productivity and speed of building and automating workflows in Pega. And when I said that, I had a number of my own internal staff go, why couldn't you have just said 20%. But I will tell you, as I sit here, looking at where I think we will be in June, I think we will be nicely past that 50%. But it's more important than that. The most important thing is not just this makes this faster. The most important thing is it stimulates the creativity of people to think differently about how they want to automate their processes and decisions. It doesn't have [ very dramatic work.]

If you think about the two...

Yes, I'm going to you next.

Well, the two -- just real quick, the 2 big risks, not just time, but the 2 big risks when you're doing something new, when you're thinking about implementing a new workload, there's 2 things. One, is to take the exact structure that you have with the system that you're mothballing and copy it, right, which is the tendency that clients have, right? Well, I have this, let's just build me that in a new -- that's a risk. The second risk is if it's something new, it takes forever, right, because you have big ideation process, you have to bring people in, then people go away, then you say, you bring everyone back, and they go where were we. I don't think -- I don't remember saying that. Like it's just the whole ideation process. And Blueprint really helps to address both of those risks and the time element, which it doesn't just -- doesn't allow you to just say copy what I have, but it also gets you so much further down the path, and that's what gen AI. That's the leverage -- leveraging gen AI certainly helps us. But it's also the content. I mean what Pat did mention is when you saw it, you'll remember when we did a use case that we had never seen in Pega before, it looked a little bit more general, right, because it was taking information on the Internet. Once we introduced a use case that we had done before, it was incredibly robust. It had like -- so the key is our differentiated content that we have as well at Pega, that actually it's not just going to the Internet, it's also the -- as Alan mentioned, the decades of us supporting and knowing exactly how processes should work for certain use cases.

So my question for you, Ken, was, will this help you accelerate ACV growth?

It's -- there's no doubt that it will shrink the selling cycle, faster time to get things into pipe, and help clients that might have 30 projects that they're not sure which one to start, get that ideation process quickly because otherwise, they're not going to look at all 30, and they're going to try to pick 2 or 3, and then they're going to go through this, could be 2 weeks, could be 90 days, process of trying to figure out exactly what that architecture diagram looks like. And we're giving them a tool that they can do all 30 in an afternoon, right, to be able to give them -- and that's if they actually have like a robust process, it would take the afternoon. It could take 2 minutes, if you just wanted to do a version 1. So to me, that's the most exciting thing that I have seen since I started at Pega, is that ability to really allow the kind of the problem to solution be compressed. And that's a big challenge for anybody selling enterprise technology, right? And we are in that bucket.

Awesome. Okay. On the regular business, any -- what's the comment on the whole Appian lawsuit at this point?

We had our appeal on November 15. We were very pleased with what we heard from the judge's comments, but we're here waiting for the decision to be written. And we have absolutely no control over when that comes out. I've been betting that it would happen in March, but it could happen in April. And I've even heard the May word, though I would not expect that. I think definitely within 60 days.

All right. Two minutes for a question or two from our audience. Yes, please.

[indiscernible] process is [indiscernible] portfolio and then [indiscernible]

Yes. So the question is how can you use existing processes in a company to figure out how you want to change it? So we acquired, about 1.5 years ago, a company in the process mining space. And the reason we did it was not to go out and try to compete with Celonis, the reason we did it is we wanted to really weave that in to Pega and the Pega tools. So we will have the ability to use the process mining as an input to what the Blueprint is able to do and really help you take but improve your common stages and steps in whatever technology you are. So that's...

[indiscernible]

I'm sorry.

[indiscernible]

I'm expecting it this year.

30 seconds.

Super quick. Just with the value that Blueprint is adding to their customers, how are you thinking that Pega is going to get value as well? Like how Pega [indiscernible] Blueprint?

So our plan is to put Blueprint on pega.com for free so that customers and prospects both can create Blueprints without charge. And if your objective is to create a Blueprint and show it to your compliance officer, go ahead and knock yourself out. Now the benefit we'll get is our advantage in being able to take a Blueprint and pull it in to create a running system. But if people begin adopting the Pega sort of design architecture as the way they describe processes and workflows in their business, I see that as only highly advantageous.

Yes, that's a win.

The way we -- yes, just to hit that point really, we monetize based on volume.

That was really short. Great.